APPLICATION BRIEF
ANSYS® Fluent® Helps Reduce Vehicle Wind Noise
ANSYS Fluent
Situation
ANSYS Fluent is a computational fluid
dynamics (CFD) software solution used
to predict flow, turbulence, heat transfer
and reactions for industrial applications
ranging from air flow over an aircraft wing
to combustion in a furnace, from bubble
columns to oil platforms, from blood flow
to semiconductor manufacturing, and from
cleanroom design to wastewater treatment
plants. Advanced solver technology provides
fast, accurate CFD results, flexible moving
and deforming meshes, and superior parallel
scalability. Additionally, Fluent has a record
of outstanding parallel scalability up to tens
of thousands of CPU cores, enabling highfidelity results in the shortest possible time.
The drivetrain, the tires on the road and the aerodynamics: these are the factors that
add noise to every driving experience. At low speed and high engine load, drivetrain
noise dominates. At low speed and low engine load, it’s where the rubber meets
the road that most noise originates. Things change at higher speeds, where the
aerodynamic noise of ground vehicles becomes dominant at driving speeds above
100 kph.
Cray® XC40™ Supercomputer
The Cray XC40 supercomputer is a massively
parallel processing architecture focused on
producing more capable HPC systems to
address a broad range of user communities. The
system leverages the combined advantages
of next-generation Aries interconnect and
Dragonfly network topology, Intel® Xeon®
processors, integrated storage solutions and
major enhancements to the Cray OS and
programming environment. The Cray® XC®
series supercomputer is a groundbreaking
architecture upgradable to 100 petaflops per
system and delivering:
• Sustained and scalable application
performance
• Tight HPC optimization and integration
• Production supercomputing
• Investment protection – upgradability by
design
• User productivity
Interior noise isn’t just an annoyance that makes it difficult for passengers to converse
or enjoy their music. It can also be a hazard because it contributes to driver fatigue,
particularly on long trips. To stay ahead of the competition and meet consumers’
ever-increasing expectations for a comfortable ride, automotive OEMs are using
ANSYS computational fluid dynamics (CFD) simulation software to tackle the problem
of wind noise. Used early in the product development process, the software helps
designers predict wind noise and then alter their designs to reduce it.
Challenge
When a vehicle – in this application, the Alfa Romeo Giulietta – is driven at high
speed, the side-view mirrors are a major source of broadband (up to about 5,000 Hz)
wind noise. The low-frequency range is dominated by large alternating vortices shed
from the top and bottom edges of each mirror. This highly energetic wake, which
resembles a Von Karman vortex street, strikes the side windows, creating noise.
Another source of wind noise at the side windows is the A-pillar vortex, which
originates at the pillar between the windshield and the side window. A quantitative
description of the wall pressure fluctuations on the side window requires knowledge
of the fine details of the flow, which can be achieved only through accurate prediction
of complex flow features, including time-dependent vortex shedding, large-to-small
turbulent scales, convection and fluid compressibility. Detached-eddy simulation
(DES) is a novel approach that combines the concepts of unsteady Reynolds-averaged
Navier-Stokes (URANS) and large-eddy simulation (LES) to obtain realistic solutions
for practical high-Reynolds-number flows at acceptable computational costs. Delayed
detached-eddy simulation (DDES) is a version of DES designed to ensure that attached
boundary layers are treated inside the RANS zone to avoid modelled stress depletion.
The DDES turbulence model was used to simulate side-window wall-pressure
fluctuation noise for the Alfa Romeo Giulietta, and a very accurate transient simulation
was performed using ANSYS Fluent CFD software running on a Cray® XC40™
supercomputer. To ensure accurate results, the test employed a very fine hexcore
mesh of 62 million cells to capture both large- and small-scale fluctuation.
Cray Inc.
901 Fifth Avenue, Suite 1000
Seattle, WA 98164
Tel: 206.701.2000
Fax: 206.701.2500
www.cray.com
Image courtesy of FCA Italy
©2015 Cray Inc. All rights reserved. Cray is a registered trademark of Cray Inc.
All other trademarks mentioned herein are the properties of their respective owners. 20150610
APPLICATION BRIEF
Page #2
Solution
Solving this demanding problem in a meaningful time frame requires a large number of compute cores. ANSYS Fluent is designed to
take advantage of multiple cores using its MPI-based parallel implementation. Efficiently executing in parallel across a large number of
cores depends on moving large amounts of data between the cores. The Cray XC40 system is designed to maximize the performance of
interprocessor communication, allowing the cores to spend their time computing. One of the key features of the XC40 supercomputer
is its proprietary Aries interconnect, which delivers low latency, high bandwidth and adaptive routing. This allows effective scaling for
demanding simulations like this one, regardless of other jobs running on the system.
In conjunction with the Aries interconnect, the XC40 system uses a proprietary MPI library optimized to take advantage of the unique
features of Aries. ANSYS Fluent is built with the Cray MPI library, which means it delivers exceptional parallel performance. In addition,
the Cray Linux® Environment, Cray’s Linux-based operating system, has been optimized to reduce overhead, allowing more of the cores’
cycles to be used for productive computation.
To ensure that users of ANSYS Fluent and Cray’s scalable systems benefit from the best possible performance, engineers from ANSYS
and Cray closely collaborate to continually improve the performance of Fluent on Cray systems. Guided by mutual customers with
demanding problems, the teams have identified numerous performance improvements, which ANSYS has incorporated into Fluent
releases over the last four years.
The curve below shows the scaling performance of ANSYS Fluent 16.0 for the Alfa Romeo Giulietta aeroacoustics case.
Fluent Simulation, 62 million cells
Solutions per Day (40 Time Steps)
Fluent 16.0.0
Cray® XC40™
1,600
1,400
1,200
Solver Rating
1,000
Ideal Scaling
800
600
400
200
0
0
2,048
4,096
6,144
Number of Cores
8,192